© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 90968
    Prev
    Next
    INDEX
    Explanations

    Python packaging

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron is triggered by tokens that consist of digits (e.g. version numbers, numeric codes, or other purely numeric strings).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    wstring
    -0.74
    endenza
    -0.73
     sinking
    -0.72
    quester
    -0.70
     prey
    -0.68
     губ
    -0.67
    avat
    -0.66
     dam
    -0.66
     تغی
    -0.66
    湖
    -0.66
    POSITIVE LOGITS
     PEP
    1.02
    extras
    1.02
    Wheel
    1.02
     wheel
    0.98
     WHEEL
    0.97
     Wheel
    0.96
    egg
    0.96
     pip
    0.94
     eggs
    0.89
     Requirement
    0.88
    Activations Density 0.097%

    No Known Activations